EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Integrating Feature Selection and Deep Learning: A Hybrid Approach for Smart Agriculture Applications

Ali Roman, Md Mostafizer Rahman, Sajjad Ali Haider, Tallha Akram, Syed Rameez Naqvi

Algorithms · 12.04.2025 · 18 (4) · 222

DOI: 10.3390/a18040222 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Algorithms, online erschienen am 12.04.2025. Verfasst von Ali Roman und 4 weiteren Autor:innen. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This research tackles the critical challenge of achieving precise and efficient feature selection in machine learning-based classification, particularly for smart agriculture, where existing methods often fail to balance exploration and exploitation in complex, high-dimensional datasets. While current approaches, such as standalone nature-inspired optimization algorithms, leverage biological behaviors for feature selection, they are limited by their inability to synergize diverse strategies, resulting in suboptimal performance and scalability. To address this, we introduce the Hybrid Predator Algorithm for Classification (HPA-C), a novel hybrid feature selection algorithm that uniquely integrates the framework of a nature-inspired feature selection technique with position update equations from other algorithms, harnessing diverse biological behaviors like echolocation, foraging, and collaborative hunting. Coupled with a custom convolutional neural network (CNN), HPA-C achieves superior classification accuracy (98.6–99.8%) on agricultural datasets (Plant Leaf Diseases, Weed Detection, Fruits-360, and Fresh n Rotten) and demonstrates exceptional adaptability across diverse imagery applications.

Zitieren

APA

Roman, A., Rahman, M. M., Haider, S. A., Akram, T., & Naqvi, S. R. (2025). Integrating Feature Selection and Deep Learning: A Hybrid Approach for Smart Agriculture Applications. Algorithms, 18(4), 222. https://doi.org/10.3390/a18040222

Vancouver

Roman A, Rahman MM, Haider SA, Akram T, Naqvi SR. Integrating Feature Selection and Deep Learning: A Hybrid Approach for Smart Agriculture Applications. Algorithms. 2025;18(4):222. doi:10.3390/a18040222

BibTeX
@article{Roman2025,
  author = {Roman, Ali and Rahman, Md Mostafizer and Haider, Sajjad Ali and Akram, Tallha and Naqvi, Syed Rameez},
  title = {Integrating Feature Selection and Deep Learning: A Hybrid Approach for Smart Agriculture Applications},
  journal = {Algorithms},
  year = {2025},
  volume = {18},
  number = {4},
  pages = {222},
  doi = {10.3390/a18040222}
}
Kostenlos starten →